You are on page 1of 23

digital

anthropology
Edited by Heather A. Horst and Daniel Miller
-2-
Rethinking Digital Anthropology
Tom Boellstorff
If there is to be such a thing as digital anthropology, we must carefully consider
both component terms constituting that promising phrase. In this chapter I respond
to a staggering analytical imbalance: while anthropology has long been subjected to
forms of critique-postcolonial, reflexive and poststructuralist, among others-to
date the notion of the digital has been met by a profound theoretical silence. For the
most part, as I have noted elsewhere, it 'does little more than stand in for "compu-
tational" or "electronic"' (Boellstorff 2011: 514). However, if digital is but a place-
holder, simply marking interest in that which you plug in to run or recharge, the
enterprise of digital anthropology is doomed to adjectival irrelevance from the out-
set. Technology is now ubiquitous worldwide, and few, if any, future fieldwork pro-
jects could ever constitute 'ethnography unplugged'. If digital is nothing more than
a synonym for Internet-mediated, then all anthropology is now digital anthropology
in some way, shape or form. Should we allow to take root a conception of digital
anthropology founded in an uninformed notion of the digital, we thus short-circuit
our ability to craft research agendas and theoretical paradigms capable of grappling
effectively with emerging articulations of technology and culture.
This highly consequential project of rethinking the digital with regard to digital
anthropology is my analytical goal in this chapter. In Part 1, I begin by addressing an
issue with foundational implications for what we take digital anthropology to mean:
the relationship between the virtual (the online) and the actual (the physical or offiine).
1
This relation has pivotal ontological, epistemological and political consequences: it
determines what we take the virtual to be, what we take knowledge about the virtual to
entail and what we understand as the stakes ofthe virtual for social justice. I focus on
the greatest negative ramification of an undertheorized notion of the digital: the mis-
taken belief that the virtual and the actual are fusing into a single domain. In Part 2, I
engage in the classic anthropological practice of close ethnographic analysis, through
case studies drawn from two early days of my research in the virtual world Second
Life. In Part 3, I link the theoretical discussion ofPart 1 with the ethnographic discus-
sion of Part 2-another classical anthropological practice, that of 'tack[ing] between
the most local of local detail and the most global of global structure in such a way as
to bring them into simultaneous view' (Geertz 1983: 68).
-39-
40 Digital Anthropology
The linchpin of my analysis will be an argument for treating the digital not as an
object of study, but as a methodological approach, founded in participant observa-
tion, investigating the virtual and its relationship to the actual. I thereby suggest
that digital anthropology is not analogous to, say, medical anthropology or legal
anthropology. The parallel to these would be virtual anthropology (Boellstorff 2008:
65). Digital anthropology is a technique, and thus a domain of study only indirectly.
It is an approach to researching the virtual that permits addressing that object of
study in its own terms (in other words, not as merely derivative of the offline), while
keeping in focus how those terms always involve the direct and indirect ways online
sociality points at the physical world and vice versa. Crucially, it is predicated on
participant observation. An alarming number of researchers of the online claim to do
ethnography when their methods involve interviewing in isolation or in conjunction
with other elicitation methods, such as a survey. But while such elicitation methods
can produce valuable data, a research project using only such elicitation methods
is not ethnographic (though it may be qualitative). Just saying something is ethno-
graphic does not make it so.
In short, while some will likely equate digital anthropology with virtual anthro-
pology, I here wish to consider a more focused conception, one inspired by originary
of the digital and that offers specific methodological benefits for studying
onlme culture. To foreshadow the crux of my argument, I develop a notion of the
digital that hearkens back to its original meaning of digits on a hand? Rather than a
diffuse notion of the digital as that which is merely electronic or online, this opens
the door to a radically more robust conceptual framework that contains two key
elements. The first is a foundational appreciation for the constitutive role of the gap
between the virtual and actual (like the gaps between 'digits' on a hand). This reso-
nates with the dialectical understanding of the digital developed by Miller and Horst
in introduction to this volume. The second element of this digital framework,
drawmg from the etymology of index as 'forefinger', is a whole set of theoretical
resources for understanding the indexical relationships that constantly co-constitute
both the virtual and actual. I thus push toward an indexical theory for understanding
how the virtual and the actual 'point' at each other in social practice.
Part 1: Challenging the Notion of Blurring
Before turning to this theory of digital anthropology and the ethnographic encounters
that inspired it, it is imperative to first identify the core problem to which a more
carefully articulated notion of digital anthropology can respond. This is the idea
that we can no longer treat the virtual and the physical as distinct or separate. It lies
beyond the scope of this chapter to catalogue examples of scholars framing the study
of the online in this manner, as this is not a review essay or even a critique as such.
3
In her insightful overview of the ethnography of digital media, E. Gabriella Coleman
nicely summed up this perspective when noti
tual worlds, 'the bulk of this work, however,
ies between off-line and online contexts' (Cc
captured the sense that 'sharp boundaries' e
arly conceits that falsely separate online an
cally consequential gaps that constitute the
boundaries are real, and therefore vital topic
While less evident in this particular quota
1
the online and offline as separate--despite t
on how you define 'separate' --encodes a his
tion to blurring or fusion. Such presumption
the virtual and actual mischaracterize the cal
online.
4
For instance, Vili Lehdonvirta has ell
is 'based on a dichotomous "real-virtual" per:
sustain this view that scholars have detached
only through a sociology of the obvious-nc
game like World of Warcraft often seek to
1
and time zone in which they reside' (2), as
aware of this fact. Lehdonvirta correctly col
worlds] side-by-side with spheres of
them using the same conceptual tools' (2) ar
aries altogether and let research lose its foe
ies based on technological distinctions' (9)
assumption that virtual worlds are artificial
as family, work or golf are somehow not 2
tinctions are central to the human condition
tially human endeavour. To presume othef1
authenticity', which, as Miller and Horst nc
iota more mediated by the rise of digital te
Thus, the most significant danger lie!
the three-part narrative of movement em1
dichotomies and blurring: an originary se
tion. This narrative is a teleology insofar
ing nonseparation of the virtual and the
language of 'the end of the virtual/real d
tentions of an end times represent not ju
so often appear as articles of faith with
resemble nothing so much as the Christi:
separation of God from Man in Eden res
2011).
7
This speaks to pervasive Judea
dualism of flesh and spirit' that have st
quiry (Sahlins 1996: 400).
~ an argument for treating the digital not as an
:al approach, founded in participant observa-
ts relationship to the actual. I thereby suggest
>gous to, say, medical anthropology or legal
uld be virtual anthropology (Boellstorff2008:
te, and thus a domain of study only indirectly.
rirtual that permits addressing that object of
not as merely derivative of the offline), while
ys involve the direct and indirect ways online
and vice versa. Crucially, it is predicated on
Jmber of researchers ofthe online claim to do
ve interviewing in isolation or in conjunction
a survey. But while such elicitation methods
project using only such elicitation methods
qualitative). Just saying something is ethno-
.ate digital anthropology with virtual anthro-
ocused conception, one inspired by originary
;pecific methodological benefits for studying
: of my argument, I develop a notion of the
l meaning of digits on a hand.
2
Rather than a
ch is merely electronic or online, this opens
mceptual framework that contains two key
>reciation for the constitutive role of the gap
gaps between 'digits' on a hand). This reso-
ofthe digital developed by Miller and Horst
e second element of this digital framework,
iS 'forefinger', is a whole set of theoretical
2/ relationships that constantly co-constitute
oward an indexical theory for understanding
each other in social practice.
Blurring
~ r o p o l o y and the ethnographic encounters
identify the core problem to which a more
thropology can respond. This is the idea
d the physical as distinct or separate. It lies
gue examples of scholars framing the study
a review essay or even a critique as such. 3
phy of digital media, E. Gabriella Coleman
Rethinking Digital Anthropology 41
nicely summed up this perspective when noting that, with regard to research on vir-
tual worlds, 'the bulk of this work, however, continues to confound sharp boundar-
ies between off-line and online contexts' (Coleman 2010: 492). Coleman's phrasing
captured the sense that 'sharp boundaries' are to be avoided-that they are schol-
arly conceits that falsely separate online and offline contexts rather than ontologi-
cally consequential gaps that constitute the online and offline. In fact, these sharp
boundaries are real, and therefore vital topics for anthropological inquiry.
While less evident in this particular quotation, the sense that one can no longer see
the online and offline as separate-despite the obvious fact that they are, depending
on how you define 'separate' --encodes a historical narrative that moves from separa-
tion to blurring or fusion. Such presumptions of an impending convergence between
the virtual and actual mischaracterize the careful work of earlier ethnographers of the
online.
4
For instance, Vili Lehdonvirta has claimed that much virtual-world scholarship
is 'based on a dichotomous "real-virtual" perspective' (Lehdonvirta 20 l 0: 2).
5
He could
sustain this view that scholars have detached virtual worlds from 'the rest of society' (2)
only through a sociology of the obvious-noting, for instance, that players of an online
game like World of Warcraft often seek to play with persons 'based on the continent
and time zone in which they reside' (2), as if World ofWarcraft researchers were not
aware of this fact. Lehdonvirta correctly concluded that 'scholars should place [virtual
worlds] side-by-side with spheres of activity such as family, work or golf, approaching
them using the same conceptual tools' (2) and that 'the point is not to give up on bound-
aries altogether and let research lose its focus, but to avoid drawing artificial boundar-
ies based on technological distinctions' (9). What needs questioning is Lehdonvirta's
assumption that virtual worlds are artificial boundaries, while spheres of activity such
as family, work or golf are somehow not artificial.
6
At issue is that technological dis-
tinctions are central to the human condition: artifice, the act of crafting, is a quintessen-
tially human endeavour. To presume otherwise sets the stage for the 'principle of false
authenticity', which, as Miller and Horst note, occludes the fact that 'people are not one
iota more mediated by the rise of digital technologies' (this volume: 11-12).
Thus, the most significant danger lies not in scholarly misrepresentation but in
the three-part narrative of movement embedded in these concerns over authenticity,
dichotomies and blurring: an originary separation, a coming together and a reunifica-
tion. This narrative is a teleology insofar as there is a defining endpoint: the impend-
ing nonseparation of the virtual and the actual, often presented in the apocalyptic
language of 'the end of the virtual/real divide' (Rogers 2009: 29). Indeed, such con-
tentions of an end times represent not just a teleology but a theology-because they
so often appear as articles of faith with no supporting evidence, and because they
resemble nothing so much as the Christian metaphysics of incarnation, of an original
separation of God from Man in Eden resolved in the Word made flesh (Bedos-Rezak
2011 ).
7
This speaks to pervasive Judea-Christian assumptions of 'the antagonistic
dualism of flesh and spirit' that have strongly shaped dominant forms of social in-
quiry (Sahlins 1996: 400).
42 Digital Anthropology
Without cataloguing further examples of these narratives that the online and off-
line are becoming blurred, it is important to note their persistence despite the fact
that this transcendental understanding of the virtual is clearly wrong: the virtual is
as profane as the physical, as both are constituted 'digitally' in their mutual relation-
ship. This language of fusion undermines the project of digital anthropology; it is
an eschatological narrative, invoking an end times when the virtual will cease to be.
This recalls how some scholars of the online seem unable to stop referring to the
physical as the 'real', even though such inaccurate phrasing implies that the online is
unreal-de legitimizing their field of study and ignoring how the virtual is immanent
to the human. The persistence of such misrepresentations underscores the urgent
need for rethinking digital anthropology.
Some readers may have recognized the homage at play in my phrase 'rethinking
digital anthropology'.
8
In 1961, the eminent British anthropologist Edmund Leach
published the essay 'Rethinking Anthropology'. In it, he chose a fascinating analogy
to justify anthropological generalizations:
Our task is to understand and explain what goes on in society, how societies work. If
an engineer tries to explain to you how a digital computer works he doesn't spend his
time classifying different kinds of nuts and bolts. He concerns himself with principles,
not with things. He writes out his argument as a mathematical equation of the utmost
simplicity, somewhat on the lines of: 0+1 = 1; 1+1 = 10 ... [the principle is that] com-
puters embody their information in a code which is transmitted in positive and negative
impulses denoted by the digital symbols 0 and 1. (Leach 1961: 6-7)
Leach could have not have predicted the technological transformations that now
make digital anthropology possible. Nonetheless, we can draw two prescient in-
sights from his analysis. First, 39 years after Bronislaw Malinowski established in
Argonauts of the Western Pacific that 'the essential core of social anthropology is
fieldwork' (Leach 1961: 1; see Malinowski 1922), Leach emphasized that anthro-
pologists must attend to the 'principles' shaping everyday life. Second, to illustrate
these principles, Leach noted the centrality of gaps to the digital: even a computer of
nuts and bolts depends on the distinction between 0 and 1.
Leach's observations anticipate my own argument. The persistence of narratives
bemoaning the distinction between the physical and the online miss the point-
literally 'miss the point', as my discussion of indexicality in Part 3 will demonstrate.
The idea that the online and offline could fuse makes as much sense as a semiotics
whose followers would anticipate the collapsing of the gap between sign and refer-
ent, imagining a day when words would be the same thing as that which they denote.
9
Clearly, we need a range of conceptual resources to theorize traffic across consti-
tutive gaps; allow me to provide an example from my research on sexuality. In my
studies of men who use the Indonesian term gay to describe their sexualities, I sought
a framework that would not lead me to presume these men were becoming the same
Re
as Western gay men. I found such a resource fn
often discovers via an ethnographic
tried to ban the dubbing of foreign
language with the justification that to see Sl
cause Indonesians to lose the ability to tell
culture began (Oetomo 1997; see
What is interesting about dubbmg lS Its e)
a
gap I
n a dubbed movie-say, an Ita
across .
. l' s of the Italian actors will never <
movmg 1p .
m
embers of an audience wtllleave the th<
no r d
pected not a failure so long as the Ips an "
under:tanding can take place.\0 Inspired by
vel oped a notion of' dubbing to avOl
d endpomt for 1
tity represented the assume . .
gay men dub Western gay sexuahtles. They
is shaped by the English term ga_
term gay . .
their subjectivities are not merely denvatlv<
The notion of dubbing culture helped 1
Western sexualities were bh
involves movement across gaps.
help avoid any assumption that the vtrtual
I Part 3 I discuss what such a rethought r
:ar such 'a rethinking to apply to digital an
divorced from questions of method. In Pa
trajectory of this argument to reflect how
graphic engagement. i.s not.a
of anthropological inqmry IS takmg.ethn :
not just data in service of preconceiVed P'
Part 2: Two Days in My Early
Given the scope of this chapter, l cannot d
Life.12 Briefly, Second Life is a virtual W(
computer programme through :he Intern
avatar body and can interact With other 1
at the same time; the virtual world rema
ff because it is housed in the 'cloud'' <
o 'when I first joined Second Life
and were provided a small plot of virtt
had been initially allocated and
this chapter in 2011' to get myself mtc
s of these narratives that the online and off-
nt to note their persistence despite the fact
f the virtual is clearly wrong: the virtual is
nstituted 'digitally' in their mutual relation-
es the project of digital anthropology; it is
end times when the virtual will cease to be.
mline seem unable to stop referring to the
1accurate phrasing implies that the online is
Y and ignoring how the virtual is immanent
misrepresentations underscores the urgent
te homage at play in my phrase 'rethinking
1ent British anthropologist Edmund Leach
~ l o g y In it, he chose a fascinating analogy
bat goes on in society, how societies work. If
l digital computer works he doesn't spend his
r1d bolts. He concerns himself with principles,
ent as a mathematical equation of the utmost
I= 1; 1+1 = 10 ... [the principle is that] com-
~ which is transmitted in positive and negative
I and I. (Leach 1961 : 6-7)
e technological transformations that now
1etheless, we can draw two prescient in-
tfter Bronislaw Malinowski established in
te essential core of social anthropology is
ski 1922), Leach emphasized that anthro-
haping everyday life. Second, to illustrate
V of gaps to the digital: even a computer of
'etween 0 and I.
rn argument. The persistence of narratives
,hysical and the online miss the point-
' of indexicality in Part 3 will demonstrate.
fuse makes as much sense as a semiotics
psing of the gap between sign and refer-
the same thing as that which they denote.9
esources to theorize traffic across consti-
le from my research on sexuality. In my
gay to describe their sexualities, I sought
ume these men were becoming the same
Rethinking Digital Anthropology 43
as Western gay men. I found such a resource from the kind of unexpected quarter one
often discovers via an ethnographic approach. I learned that the Indonesian state had
tried to ban the dubbing of foreign television shows and movies into the Indonesian
language with the justification that to see 'Sharon Stone speak Indonesian' would
cause Indonesians to lose the ability to tell where their culture ended and Western
culture began (Oetomo 1997; see Boellstorff2005).
What is interesting about dubbing is its explicit predication on meaning-making
across a gap. In a dubbed movie-say, an Italian movie dubbed into Japanese-the
moving lips of the Italian actors will never exactly match the Japanese voices. Yet
no members of an audience will leave the theatre because of this mismatch: it is ex-
pected, not a failure so long as the lips and voices are close enough in synch so that
understanding can take place.
10
Inspired by these antiteleological implications, I de-
veloped a notion of 'dubbing culture' to avoid a narrative in which Western gay iden-
tity represented the assumed endpoint for homosexualities worldwide. Indonesian
gay men dub Western gay sexualities. They are perfectly aware that the Indonesian
term gay is shaped by the English term gay, yet they are also perfectly aware that
their subjectivities are not merely derivative of the West.
The notion of dubbing culture helped me avoid assuming that Indonesian and
Western sexualities were converging or blurring and underscored how all semiosis
involves movement across gaps. Similarly, extending the notion of the digital can
help avoid any assumption that the virtual and actual are converging or blurring.
11
In Part 3, I discuss what such a rethought notion of the digital might entail and how,
for such a rethinking to apply to digital anthropology, questions of theory cannot be
divorced from questions of method. In Part 2, I tum to two case studies: I want the
trajectory of this argument to reflect how my thinking has emerged through ethno-
graphic engagement. This is not a detour, digression or mere illustration; a hallmark
of anthropological inquiry is taking ethnographic work as a means to develop theory,
not just data in service of preconceived paradigms.
Part 2: Two Days in My Early Second Life
Given the scope of this chapter, I cannot devote much space to background on Second
Life.
12
Briefly, Second Life is a virtual world-a place of human culture realized by a
computer programme through the Internet. In a virtual world, you typically have an
avatar body and can interact with other persons around the globe who are logged in
at the same time; the virtual world remains even as individuals shut their computers
off, because it is housed in the 'cloud', on remote servers.
When I first joined Second Life on 3 June 2004, residents paid a monthly fee
and were provided a small plot of virtual land. In February 2005, I sold the land I
had been initially allocated and moved to another area. However, at the time I write
this chapter in 2011, to get myself into an ethnographic frame of mind, in another
44 Digital Anthropology
Figure 2.1. The land where my first home in Second Life once stood.
window on my computer I have gone into Second Life and teleported back to the
exact plot of virtual land where my original home once stood in 2004. At this mo-
ment-late morning according to my California time-there are no avatars nearby.
The large house that once stood here, my first experiment at building in Second Life
disappeared long ago, and nary a virtual nail remains of my prior labour. But
at my old land's little patch of coastline, I think I can still make out the remnants of
my terraforming, my work to get the beach to slope into the water just so, in order to
line up with the view of the distant shore to the east. Even in virtual worlds, traces of
history endure (Figure 2.1 ).
The current owners of my onetime virtual homestead have not built a new house
to replace the one I once crafted; instead, they have made the area into a wooded
parkland. To one side, swings rock to and fro with automated animations, as if bear-
ing unseen children. On the other side, at the water's edge, a dock invites repose.
In the centre, near where the living room of my old home was located, there now
stands a great tree, unlike any I have ever seen in Second Life. Its long branches
slope gracefully up toward the bright blue virtual sky. One branch, however, snakes
out horizontally for some distance; it contains an animation allowing one's avatar to
stretch out, arms folded behind one's head and feet swinging in the digital breeze. So
here on this branch, where my first Second Life home once stood, my virtual self will
sit as I reflect on those first days of virtual fieldwork (Figure 2.2).
In what follows, I recount hitherto unpublished fieldwork excerpts from two con-
current days early in my research. (Second Life at this time had only text communica-
tion, which I have edited for concision. As is usual in ethnographic writing, to protect
Figure 2.2. At rest in the virtual tree.
confidentiality all names are pseudonyms.) 1
thy; it is unlikely anyone else bothered to re(
traces of broader meaning that point to ware
Day 1: A Slow Dance for Science
At 12:28 p.m. on 30 June 2004, I
California, and turned on my computer. I
Second Life in my recently constructed b
tree seven years later as I write this nam
fieldwork, I left my virtual home and tel'
of Susan, who was already at the club wit
this point Second Life was quite small an'
the featured attraction was ice skating; the
ice skates were available on the walls to a
skates and they appeared in a box; if yot
you would end up wearing the box on yc
residents were new to the virtual
getting her skates to work, and Sam and l
Sam: Susan, take them off your :
Sam: put them onto the ground
Second Life and teleported back to the
home once stood in 2004. At this mo-
. time-there are no avatars nearby.
experiment at building in Second Life,
1 remains of my prior labour. But looking
I can still make out the remnants of
slope into the water just so, in order to
east. Even in virtual worlds, traces of
homestead have not built a new house
have made the area into a wooded
with automated animations, as if bear-
water's edge, a dock invites repose.
my old home was located, there now
seen in Second Life. Its long branches
al sky. One branch, however, snakes
an animation allowing one's avatar to
feet swinging in the digital breeze. So
home once stood, my virtual self will
leldwo1rk (Figure 2.2).
fieldwork excerpts from two con-
at this time had only text communi ca-
in ethnographic writing, to protect
Rethinking Digital Anthropology 45
Figure 2.2. At rest in the virtual tree.
confidentiality all names are pseudonyms.) None of these interactions were notewor-
thy; it is unlikely anyone else bothered to record them. Yet in each case I encountered
traces of broader meaning that point toward rethinking digital anthropology.
Day 1: A Slow Dance for Science
At 12:28 p.m. on 30 June 2004, I walked into my home office in Long Beach,
California, and turned on my computer. I 'rezzed' (that is, my avatar appeared) in
Second Life in my recently constructed house, right where my avatar will sit in a
tree seven years later as I write this narrative. But on this day, only a month into
fieldwork, I left my virtual home and teleported to a dance club at the suggestion
of Susan, who was already at the club with her friends Sam, Richard and Becca. At
this point Second Life was quite small and there were only a few clubs. At this club
the featured attraction was ice skating; the club had been decked out with a rink, and
ice skates were available on the walls to attach to your avatar. In fact you bought the
skates and they appeared in a box; if you did not know how to do things correctly,
you would end up wearing the box on your head, not the skates on your feet. Most
residents were new to the virtual world's workings; Susan was having a hard time
getting her skates to work, and Sam and Richard were helping as best they could:
Sam: Susan, take them off your head lol [laugh out loud]
Sam: put them onto the ground
46 Digital Anthropology
Susan:
Susan:
Susan:
Richard:
Richard:
Richard:
Susan:
Richard:
thanks
hehe, I'm new to this game
have I got them on?
click on the box on your head and choose edit
then click the 'more' button
then 'content' and you'll see them
I have the skateson ... I think I do anyway
she has the box on her head
Susan (and others) continued to have trouble using the skates. In the meantime, I had
managed to figure it out and was soon skating near Becca, who saw from my profile
that I was an ethnographer:
Becca:
Richard:
Susan:
Becca:
Tom:
Becca:
Susan:
Becca:
Sam:
Becca:
Richard:
Susan:
Sam:
Susan:
Becca:
Becca:
Susan:
IM [instant message]:
Tom would you like to slow dance?
they [the skates) are still in the box I believe
But I can't see it [the box] on my head
for science
how do you do it?
lol
he he
urn ... not sure
I don't see a box on her head.
he he
I do
So is it on my head then or not?
So Susan ... you get a set of skates in a box?
hehe, I think that might work
oh there we go
lol
Yeah, I got them from the box, moved them into my in-
ventory and then put them on
Becca: just don't put your hand up my skirt ... hehe
Despite the fact that I have edited this conversation for the sake of brevity, the ethno-
graphic detail in this excerpt alone could take many pages to properly analyse, and it
illustrates the kinds of data obtainable from participant observation that could not be
acquired via interviews or other elicitation methods. I will note just six insights we
can glean from this fieldwork encounter.
First, residents worked together to educate each other rather than relying on the
company that owns Second Life or some kind of instruction manual.
Second, gender seems to be shaping the interaction: it is largely men advising
women. Since everyone knows that physical-world gender might not be aligning
with virtual-world gender, this has implications for social constructions of gender.
Third, during this period when Second
the introduction of voice in 2007, chat rem
parse conversations in which there were n
instance, Sam asked Susan, 'you get a set <
three lines later, after first answering, 'I thin
ferent thread of conversation.
Fourth, when Becca made a slightly risq
hand up my skirt'), she switched to an inst
visible to no one besides myself. This appa
early in my research that I should attend no
their modality 'shou
but to avatars at a greater distance) and insta
groups of residents. These various modalit
linguistic interest in codeswitching but car
between different technological modalities <
Fifth, these insights (and many more) hac
Peer education, the impact of gender norms
be ascertained and the existence of multiply
were not unique to this interaction, to Seco1
an awareness of relevant literatures proved l
Sixth, this encounter underscored how 1
The fact that I was participating in Second L
impediment; rather, it made the research mor
illustrated the practice of participant observ2
Day 2: Here and There
On 1 July 2004, one day after my slow dane
again to conduct fieldwork, appearing as usu
instantaneously to another part of the virtua
path. In the distance I saw three avatars, Rot
Robert:
Karen:
Timothy:
Tom:
Karen:
Karen:
Robert:
Tom:
Robert:
Karen:
Robert:
Why, hello!
Hi Tom
Hi tom
Hello! I'm your neighbor do\\
Ahh cool
Sorry for all the mayhem here
Hope the hoopla hasn't been a
What hoopla are you talking a
Hee hee
rofi [rolling on the floor laugh
just asking for it!
md choose edit
em
do anyway
using the skates. In the meantime, I had
g near Becca, who saw from my profile
ike to slow dance?
!lre still in the box I believe
[the box] on my head
on her head.
ad then or not?
get a set of skates in a box?
:might work
1 from the box, moved them into my in-
put them on
put your hand up my skirt ... hehe
sation for the sake of brevity, the ethno-
many pages to properly analyse, and it
articipant observation that could not be
ethods. I will note just six insights we
te each other rather than relying on the
d of instruction manual.
interaction: it is largely men advising
al-world gender might not be aligning
ons for social constructions of gender.
Rethinking Digital Anthropology 47
Third, during this period when Second Life had only text chat (and even after
the introduction of voice in 2007, chat remained common), residents had learned to
parse conversations in which there were multiple threads of overlapping talk. For
instance, Sam asked Susan, 'you get a set of skates in a box?' and Susan answered
three lines later, after first answering, 'I think that might work', in reference to a dif-
ferent thread of conversation.
Fourth, when Becca made a slightly risque comment to me ('just don't put your
hand up my skirt'), she switched to an instant message, meaning that this text was
visible to no one besides myself. This apparently trivial practice helped me realize
early in my research that I should attend not just to the content of statements but to
their modality of articulation-'chat', 'shout' (text that, like chat, is publicly visible
but to avatars at a greater distance) and instant messages sent both to individuals and
groups of residents. These various modalities of articulation link to long-standing
linguistic interest in codeswitching but can also take forms of 'channelswitching'
between different technological modalities of communication (Gershon 2010a).
Fifth, these insights (and many more) had precedents and contemporary parallels.
Peer education, the impact of gender norms even when physical-world gender cannot
be ascertained and the existence of multiply threaded and multimodal conversations
were not unique to this interaction, to Second Life or even to virtual worlds. Thus,
an awareness of relevant literatures proved helpful in analysing these phenomena.
Sixth, this encounter underscored how the ethnographer is not a contaminant.
The fact that I was participating in Second Life culture without deception was not an
impediment; rather, it made the research more scientific. My 'slow dance for science'
illustrated the practice of participant observation, online and offline.
Day 2: Here and There
On 1 July 2004, one day after my slow dance for science, I logged into Second Life
again to conduct fieldwork, appearing as usual in my house. Rather than teleporting
instantaneously to another part of the virtual world, I walked down a nearby paved
path. In the distance I saw three avatars, Robert, Karen and Timothy:
Robert:
Karen:
Timothy:
Tom:
Karen:
Karen:
Robert:
Tom:
Robert:
Karen:
Robert:
Why, hello!
Hi Tom
Hi tom
Hello! I'm your neighbor down the road
Ahh cool
Sorry for all the mayhem here, I have crazy friends
Hope the hoopla hasn't been a problem
What hoopla are you talking about?
Hee hee
rofl [rolling on the floor laughing] whew
just asking for it!
48 Digital Anthropology
Timothy:
Karen:
Tom:
Karen:
Karen:
Tom:
Karen:
Karen:
Tom:
Karen:
Karen:
Tom:
whew
Oh the a vie [avatar] launch game we had ... the explosions, lap dances
Whatever it is, is hasn't bothered me!
Very good
So which way down the road are you?
To my right
Ah very good
Got a house, or doing something else there?
Just got a place for now
cool
Gonna tum this into a small boutique
cool!
Already from the discussion, I had noted how copresence in a virtual neighbourhood
could help shape online community: place matters online. Karen then changed the
subject:
Karen:
Karen:
Karen:
Tom:
Karen:
Tom:
Karen:
Tom:
Tom:
Karen:
Karen:
Tom:
Karen:
Timothy:
Timothy:
Timothy:
Tom:
Karen:
Timothy:
Karen:
Robert:
Karen:
Timothy:
wow Tom, reading your profile here.
very interesting
urn ... Indonesia, really?
Yep! Cool place. Not cool really, hot and humid, but fun.
lol how'd you end up over there?
Random life events, backpacking there after college & meeting
people
that's gotta be quite interesting I imagine
very!
is that your glowing dance floor over there to my left?
nope, no clue who it's for
a little bright
there's a lot ofbuilding right now in this area! It's cool-every day
the landscape is transformed
yes, a lot of this land was just released
happens in new areas
finally got a house on one side of mine
mini tower going in behind
laugh
lol
as long as they don't cut off my view
they screwed up my view in Shoki [region]
Yeah, its just sad.
even though he said he wouldn't
think I am safe there
After a brief discussion of my po:
once again to virtual place. In m
across a virtual landscape. Encou
place to virtual worlds (see Boelh
tiple avatars, and I asked about Tl
explored:
Tom:
Karen:
Robert:
Timothy:
Karen:
Karen:
Tom:
Timothy:
Tom:
Karen:
Robert:
Tom:
Timothy:
Karen:
Karen:
Timothy:
Robert:
Timothy:
Robert:
Karen:
Karen:
do you play more 1
did that in The Si
here.
no, not here, in TS
Never saw the Sin
I never tried TSO
Didn't miss shit
so you missed h ~
Yes, I missed The1
I remember that
Was it more like S
Very much like 1
be pg13
Step ford Disney\
Is it still around?
and not quite as o
yes, Stepford Dis1
but there's still a 1
but it has its nice
Better chat, great
Meeting Karen b ~
Card games!
yes, I met both yc
the horizon is cle;
This section of the discussion
shaped by previous and sometin
influenced not only how the user
(for instance, Karen first met Rc
how other virtual worlds shape<
to conduct fieldwork in these otl
is certainly useful given the ap
a virtual diaspora that moves a
2009). However, it was clearly I
without visiting them personall)
1me we had ... the explosions, lap dances
ered me!
. are you?
ting else there?
)Outique
w copresence in a virtual neighbourhood
matters online. Karen then changed the
lle here.
ally, hot and humid, but fun.
ere?
acking there after college & meeting
1g I imagine
)or over there to my left?
t now in this area! It's cool-every day
t released
le of mine
Rethinking Digital Anthropology 49
After a brief discussion of my positionality as a researcher, the conversation turned
once again to virtual place. In my fieldnotes I noted the importance of one's view
across a virtual landscape. Encounters like this led me to realize the importance of
place to virtual worlds (see Boellstorff2008: chap. 4). The topic then turned to mul-
tiple avatars, and I asked about The Sims Online, another virtual world I had briefly
explored:
Tom:
Karen:
Robert:
Timothy:
Karen:
Karen:
Tom:
Timothy:
Tom:
Karen:
Robert:
Tom:
Timothy:
Karen:
Karen:
Timothy:
Robert:
Timothy:
Robert:
Karen:
Karen:
do you play more than one avie at the same time? I know people who
did that in The Sims Online but it seems that would be hard to do
here.
no, not here, in TSO [The Sims Online] I did
Never saw the Sims, did I miss much?
I never tried TSO
Didn't miss shit
so you missed There altogether?
Yes, I missed There completely. What was it like?
I remember that
Was it more like Second Life than TSO?
Very much like this, but more cartoonish and everything had to
be pg13
Stepford Disney World
Is it still around?
and not quite as open
yes, Stepford Disney lol
but there's still a lot of charm to There
but it has its nice parts
Better chat, great vehicles
Meeting Karen being one of em
Card games!
yes, I met both you guys in There
the horizon is clear, not foggy like here
This section of the discussion reveals how understandings of Second Life were
shaped by previous and sometimes ongoing interaction in other virtual worlds. This
influenced not only how the users experienced Second Life, but their social networks
(for instance, Karen first met Robert and Timothy in There.com). Yet to learn about
how other virtual worlds shaped Second Life sociality, it was not necessary for me
to conduct fieldwork in these other virtual worlds. Multisited ethnographic research
is certainly useful given the appropriate research question-for instance studying
a virtual diaspora that moves across several virtual worlds (Pearce and Artemesia
2009). However, it was clearly possible to explore how other places shape a fieldsite
without visiting them personally. Indeed, in his well-known discussion ofmultisited
50 Digital Anthropology
ethnography, George Marcus was careful to note the value of 'the strategically situ-
ated (single-site) ethnography' (Marcus 1995: 110). This was an unexpected meth-
odological resonance between my research in Second Life and Indonesia: to learn
about gay identity in Indonesia, it was unnecessary to visit Amsterdam, London or
other places that gay Indonesians saw as influencing their understanding of homo-
sexual desire.
Once again, virtually embodied presence was critical to my ethnographic method.
In this one encounter, I gained a new appreciation for virtual place, the importance of
vision and 'a good view', and the impact of other virtual worlds. I mentioned none
of these three topics in my original research proposal, even though they all turned
out to be central to my conclusions. The insights were emergent, reflecting how 'the
anthropologist embarks on a participatory exercise which yields materials for which
analytical protocols are often devised after the fact' (Strathem 2004: 5-6).
Part 3: Digital Anthropology, Indexicality
and Participant Observation
These ethnographic materials highlight how the gap between online and offline is
culturally constitutive, not a suspect intellectual artefact to be blurred or erased. This
distinction is not limited to virtual worlds. For instance, Daniel Miller has noted
that for persons in Trinidad who have difficulty with physical-world relationships,
'Facebook provides an additional space for personal expression' (Miller 2011: 169).
That is, forms of expression and relationship can take place on Facebook, but the
space ofFacebook and the space ofTrinidad do not thereby collapse into each other.
You can be on Facebook without being in Trinidad, and you can be in Trinidad
without being on Facebook. Another example: in her study of breakups online, Ilana
Gershon noted that such disconnections 'are emphatically not the disconnections
between supposedly real interactions and virtual interactions. Rather, they are dis-
connections between people-the endings of friendships and romances' (Gershon
2010b: 14). These endings are both online and offline in character. To rethink digital
anthropology, we must build upon such insights to identify a common set of is-
sues that make digital anthropology cohere, and we can then explore in particular
fieldsites. This is why I now scope out from the specificities of Second Life, and
even virtual worlds, toward a theoretical and methodological framework for digital
anthropology.
lndexicality as a Core Theory for Digital Anthropology
In the introduction, I suggested that an indexical theory for understanding the re-
lationship between virtual and actual could help in rethinking digital anthropol-
ogy. Scholars of language have long noted the existence of words that lie outside
traditional notions of reference, becam
social interaction. For instance, the trutl
Letizia de Ramolino was the mother of
[I]n no way depends on who says it, bu
we try to analyze:
I am the mother ofNapoleon
We cannot assess the truth of this s n t ~
is ... we need to know, in addition to th
in which it was uttered (here, the ident
The philosopher Charles Sanders Pei1
(Levinson 1983: 57) and emphasized
to referents. To use two examples fami
a hole in a piece of metal is an index
each case, a causal relationship 'poin1
in a piece of metal does not conventic
a drawing of a bullet shape or the wor
the hole in the piece of metal refers tc
Similarly, 'the smoke does not "stan1
might be used in telling a story abou
spatia-temporally and physically, to
"meaning" from that spatia-temporal
While these examples indicate th
whole range of words are indeed ind1
including 'the demonstrative pronout
you, temporal expressions like now, 1
down, below, above' (Duranti 1997:
its meaning shifts based on the cultt
round' or 'the sun is square' can be
tion in time and place. However, I c
table is round' unless I know the con
Indexicals can be found in all humar
instance in French and German, forn
vous and du/Sie, respectively, whicl
obligatory forms of social indexical]
As noted by Duranti, indexicals
cific social realities: 'A basic proper
is dynamic. As interactants move th
ote the value of 'the strategically situ-
: 11 0). This was an unexpected meth-
1 Second Life and Indonesia: to learn
essary to visit Amsterdam, London or
llencing their understanding of homo-
as critical to my ethnographic method.
ion for virtual place, the importance of
1ther virtual worlds. I mentioned none
proposal, even though they all turned
hts were emergent, reflecting how 'the
:rcise which yields materials for which
(Strathern 2004: 5-6).
the gap between online and offline is
1al artefact to be blurred or erased. This
For instance, Daniel Miller has noted
llty with physical-world relationships,
expression' (Miller 2011: 169).
) can take place on Facebook, but the
not thereby collapse into each other.
Trinidad, and you can be in Trinidad
.1: in her study ofbreakups online, Ilana
emphatically not the disconnections
!tual interactions. Rather, they are dis-
f friendships and romances' (Gershon
a offline in character. To rethink digital
to identify a common set of is-
land we can then explore in particular
the specificities of Second Life, and
methodological framework for digital
Anthropology
ical theory for understanding the re-
help in rethinking digital anthropol-
e existence of words that lie outside
Rethinking Digital Anthropology 51
traditional notions of reference, because their meaning depends on the context of
social interaction. For instance, the truth of the sentence:
Letizia de Ramolino was the mother of Napoleon
[l]n no way depends on who says it, but simply on the facts of history. But now suppose
we try to analyze:
I am the mother of Napoleon
We cannot assess the truth of this sentence without taking into account who the speaker
is ... we need to know, in addition to the facts of history, certain details about the context
in which it was uttered (here, the identity of the speaker). (Levinson 1983: 55--{))
The philosopher Charles Sanders Peirce termed words like these 'indexical signs'
(Levinson 1983: 57) and emphasized their causal rather than symbolic relationship
to referents. To use two examples familiar to linguists, smoke is an index of fire, and
a hole in a piece of metal is an index of the bullet that passed through the metal. In
each case, a causal relationship 'points back' from the index to the referent. A hole
in a piece of metal does not conventionally symbolize a bullet in the same way that
a drawing of a bullet shape or the word bullet can stand for an actual bullet. Instead,
the hole in the piece of metal refers to the bullet causally-the bullet made the hole.
Similarly, 'the smoke does not "stand for" the fire the way in which the word fire
might be used in telling a story about a past event. The actual smoke is connected,
spatia-temporally and physically, to another, related, phenomenon and acquires
"meaning" from that spatia-temporal, physical connection' (Duranti 1997: 17).
While these examples indicate that indexical signs do not have to be words, a
whole range of words are indeed indexicals (indexical denotationals, to be precise),
including 'the demonstrative pronouns this, that, those, personal pronouns like I and
you, temporal expressions like now, then, yesterday, and spatial expressions like up,
down, below, above' (Duranti 1997: 17). For instance this is an indexical because
its meaning shifts based on the cultural context of the utterance. To say 'the sun is
round' or 'the sun is square' can be assigned a truth value regardless of my posi-
tion in time and place. However, I cannot assign a truth value to the utterance 'this
table is round' unless I know the context to which the word this can be said to point.
Indexicals can be found in all human languages, and interesting variations exist. For
instance in French and German, formal versus informal second-person pronouns (tu/
vous and du!Sie, respectively, which in English would all be translated you) mark
obligatory forms of social indexicality.
13
As noted by Duranti, indexicals are 'grounded' in spatially and temporally spe-
cific social realities: 'A basic property of the indexical context of interaction is that it
is dynamic. As interactants move through space, shift topics, exchange information,
52 Digital Anthropology
coordinate their respective orientations, and establish common grounds as well as
non-commonalities, the indexical framework of reference changes' (Hanks 1992:
53). This 'interactive emergence ofthe indexical ground' (Hanks 1992: 66) provides
the point of entree for rethinking digital anthropology in terms of indexicality. The
spatially and temporally specific social realities are no longer limited to the physical
world; the processes of moving though space and establishing common grounds can
now take place online as well as offline. Confronted with multiple embodiments,
and thus with indexical fields of reference that are multiple in a new way, we thereby
face the virtual as an emergent set of social realities that cannot be straightforwardly
extrapolated from the physical world. For instance the social intentions, emotions,
decisions and activities that take place on Facebook cannot be reduced to the phys-
ical-world activities and identities of those who participate in it, even though these
can have physical-world consequences ranging from a romance's dissolution to a
political revolution. It is possible, for example, to become a closer friend with some-
one on Facebook without meeting that person in the physical world along the way.
The reason why it is possible to rehabilitate the digital so as to transcend its com-
mon conflation with 'online' is that the concept is fundamentally linked to indexical-
ity. The etymology of index (Latin, forefinger) and digit (Latin, finger) both refer to
the embodied act of pointing-and this has momentous implications when you can
have multiple bodies and multiple fields of reference (even when there is not a clear
avatar body involved). Building upon this characteristic of the digital through the
framework of indexicality results in a far more precise notion of digital; it compels
attention to the indexical ground of virtual culture.
14
The greatest strength of an indexical perspective is that it avoids the concep-
tual danger discussed in Part 1: the idea that the gap between the virtual and actual
is headed down a teleological path to a blurring that we might celebrate or rue. It
would be nonsensical to contend that the distinction between smoke and fire might
someday vanish, that the gap between the word sun and the massive orb of gas at
the centre of our solar system might blur or that the difference between 1 and 0
might converge into a fog of 0.5s. Yet just such an absurdity is entailed by the idea
that the online and offline can no longer be separated. At issue are myriad forms of
social practice, including meaning-making, that move within virtual contexts but
also across the gap between virtual and actual-from skates on an avatar's feet to
embodied views across a virtual landscape, from a friendship in the actual world
altered though a text message to a friendship on Facebook between two people who
never physically meet.
At a broader level, the virtual and actual stand in an 'inter-indexical relationship'
(Inoue 2003: 327); it is through the general gap between them that the emerging
socialities so in need of anthropological investigation are taking form. As online
socialities grow in number, size and genre, the density and rapidity of these digital
transactions across the inter-indexical gap between virtual and actual increase expo-
nentially. Like standing back from a pointillist painting, it appears that the dots have
blurred into brush strokes. But no 1
carefully, one sees the discreteness 1
allow them to convey meaning. Thi
comes, no matter how quickly milli
stream by as well, for the computer
In setting out this idea of an ant
ment to the indexical relationships
mean to imply that virtual m e n i n ~
am not saying that digital anthropol
tal anthropology projects need to p1
provides an empirically accurate a1
rethink digital anthropology and vi1
strong linkages to context (Keane 2(
which there are multiple contexts, rr
cal precedent but no true historical ,
While a detailed examination o
chapter, we can note in passing tha
in Peirce's analysis, are ubiquitous
so central to computing cultures). r
approach to language and meaning.
language, this particular aspect of lar
making-is more indicative of vi
dimensions of language that 'cann
culture'(Silverstein 1976: 12). Whc
pology to make sense, it must mean
even the study oflntemet-mediated
in this regard involves drawing fm
and constitutive gaps. These entailn
research questions and lines of inq1
method, the topic to which I now tu
Participant Observation as the
for Digital Anthropology
Digital anthropology typically imp
not a method; it is the written pw
(to write) indicates. Rethinking digi
( 1) the theoretical frameworks wt
(3) how we engage in the research
Ethnographers of virtual socialii
are not always anthropologists, silll
l establish common grounds as well as
rk of reference changes' (Hanks 1992:
lical ground' (Hanks 1992: 66) provides
hropology in terms of indexicality. The
:ies are no longer limited to the physical
and establishing common grounds can
:onfronted with multiple embodiments,
:tt are multiple in a new way, we thereby
ealities that cannot be straightforwardly
nstance the social intentions, emotions,
:tcebook cannot be reduced to the phys-
who participate in it, even though these
from a romance's dissolution to a
le, to become a closer friend with some-
Ill in the physical world along the way.
1te the digital so as to transcend its com-
is fundamentally linked to indexical-
and digit (Latin, finger) both refer to
momentous implications when you can
eference (even when there is not a clear
of the digital through the
pre precise notion of digital; it compels


is that it avoids the concep-
the gap between the virtual and actual
ifl'ing that we might celebrate or rue. It
stinction between smoke and fire might
sun and the massive orb of gas at
pr that the difference between 1 and 0
iuch an absurdity is entailed by the idea
At issue are myriad forms of
[ that move within virtual contexts but
skates on an avatar's feet to
from a friendship in the actual world
on Facebook between two people who
tand in an 'inter-indexical relationship'
1 gap between them that the emerging
estigation are taking form. As online
he density and rapidity of these digital
tween virtual and actual increase expo-
st painting, it appears that the dots have
Rethinking Digital Anthropology 53
blurred into brush strokes. But no matter how high the resolution, when one looks
carefully, one sees the discreteness of the dots as well as the gaps of white space that
allow them to convey meaning. This recalls how no matter how fast a computer be-
comes, no matter how quickly millions of Os and 1 s stream by, millions of gaps will
stream by as well, for the computer's functioning depends on the gaps themselves.
In setting out this idea of an anthropology that is digital by virtue of its attune-
rnent to the indexical relationships constituting the virtual and the actual, I do not
mean to imply that virtual meaning-making is exclusively indexical in character. I
am not saying that digital anthropologists need to become semioticians or that digi-
tal anthropology projects need to prioritize indexicality. At issue is that indexicality
provides an empirically accurate and conceptually rich perspective from which to
rethink digital anthropology and virtual culture. This is because indexicality entails
strong linkages to context (Keane 2003), and we now grapple with a human reality in
which there are multiple contexts, multiple worlds, multiple bodies-all with histori-
cal precedent but no true historical parallel.
While a detailed examination of semiotic theory lies beyond the scope of this
chapter, we can note in passing that symbols and icons, the other two types of sign
in Peirce's analysis, are ubiquitous in online contexts (consider the icons that are
so central to computing cultures). Nor do we need to limit ourselves to a Peirceian
approach to language and meaning. But while not all dimensions of culture are like
language, this particular aspect of language--the centrality of indexicality to meaning-
making-is more indicative of virtual sociality than the structural-grammatical
dimensions of language that 'cannot really serve as a model for other aspects of
culture' (Silverstein 1976: 12). What I am suggesting is, first, that for digital anthro-
pology to make sense, it must mean more than just the study of things you plug in or
even the study oflntemet-mediated sociality and, second, that one promising avenue
in this regard involves drawing from the digital's indexical entailments ofpointing
and constitutive gaps. These entailments have theoretical consequences that suggest
research questions and lines of inquiry. They also have important consequences for
method, the topic to which I now tum.
Participant Observation as the Core Method
for Digital Anthropology
Digital anthropology typically implies 'doing ethnography' .
15
But ethnography is
not a method; it is the written product of a set of methods, as the suffix -graphy
(to write) indicates. Rethinking digital anthropology must therefore address not just
(I) the theoretical frameworks we employ and (2) the socialities we study, but
(3) how we engage in the research itself.
Ethnographers of virtual socialities work in a dizzying range of fieldsites (and
are not always anthropologists, since ethnographic methods have a long history in
54 Digital Anthropology
sociology and other disciplines). One of the greatest virtues of ethnographic meth-
ods is that researchers can adapt them to the contexts of particular fieldsites at par-
ticular periods in time. Ethnographic research online does not differ in this regard.
However, this flexibility is not boundless. A serious threat to the rigor and legitimacy
of digital anthropology is when online researchers claim to have 'done an ethnogra-
phy' when they conducted interviews in isolation, paired at most with the analysis of
blogs and other texts. Characterizing such research as ethnographic is misleading be-
cause participant observation is the core method of any ethnographic research project.
The reason for this is that methods such as interviews are elicitation methods. They
allow interlocutors to speak retrospectively about their practices and beliefs as well
as speculate about the future. But ethnographers combine elicitation methods (like
interviews and focus groups) with participant observation, which, as a method not
predicated on elicitation, allows us to study the differences between what people
say they do and what they do.
The problem with elicitation methods in isolation is that this methodological
choice surreptitiously encodes a theoretical presumption that culture is present to
consciousness. It is predicated on the belief that culture is something in people's
heads: a set of viewpoints that an interviewee can tell the researcher, to appear later
as an authoritative block quotation in the published account. Of course, persons can
often be eloquent interpreters of their cultures; as a result, interviews should be part
of any ethnographic project. But what interviews and other elicitation methods can
never reveal are the things we cannot articulate, even to ourselves. Obvious cases of
this include things that are repressed or unconscious, an insight dating back to Freud.
Language is another example. Consider a basic phonological rule like assimilation,
where for instance the n in inconceivable becomes m in impossible because p is a
bilabial plosive (made with the lips), and the nasal n assimilates to this place of ar-
ticulation. Almost no English speakers could describe this rule in an interview, even
though they use the rule hundreds of times a day in the flow of everyday speech.
Such aspects of culture are by no means limited to language and the psyche. In
particular, theorists of practice have worked to show how much of everyday so-
cial action involves tacit knowledge. Pierre Bourdieu emphasized this point when
critiquing anthropologists who speak of 'mapping' a culture: 'it is the analogy
which occurs to an outsider who has to find his way around in a foreign landscape'
(Bourdieu 1977: 3). Take any route you traverse as part of your daily routine. If there
is a staircase in your home or office, do you know how many stairs are there? The
peril is to seek a representation of such tacit knowledge via an interview, where the
informant's discourse is shaped by the framework of elicitation 'inevitably induced
by any learned questioning' (Bourdieu 1977: 18). As a result,
the anthropologist is condemned to adopt unwittingly for his own use the representa-
tion of action which is forced on agents or groups when they lack practical mastery of a
highly valued competence and have to provide themselves with an explicit and at least
semi-formalized substitute for it in the form of a repertoire of rules. (Bourdieu 1977: 2)
Elicitation not interwoven with particip
fuse representation with reality, and th
scripts or norms rather than embodied :r:
If there is one thing that ethnographE
is essential goes without saying becaus
lent, not least about itself as a tradition'
When ethnographers ask interview qm
practice. Representations are certainly s
effects. But they cannot be conflated \:
'what does friendship mean to you?' yo
takes friendship to be. That representa1
in (and influences) a cultural context.
identical to friendship in practice.
The methodological contribution of
nographers insight into practices and
obtaining nonelicited data-conversati
ments, movements though space, and
observed Second Life residents teachin
in part by learning how to skate mysel1
out of the blue, 'how do you learn in :
formal response emphasizing things t1
tail about a group of avatars learning
Participant observation allows researc
of which they were unaware during the
Some persons terming themselves
more than one occasion I have counse
raphy' but use interviews in isolation
scholar that participant observation w
never rapid: 'not unlike learning anot
patience. There are no shortcuts' (Ros
a new language overnight, or even in
to have conducted ethnographic resear
izing his or her work unless it is part ,
way the researcher could have e c o m ~
everyday practices in such a compress
Conclusion: Time and Imaginat
When I think about the exciting possil
pology, I find my mind wandering ba<
has haunted me for years despite its ap
the original McDonald's home page fr
itest virtues of ethnographic meth-
texts of particular fieldsites at par-
lline does not differ in this regard.
cts threat to the rigor and legitimacy
s claim to have 'done an ethnogra-
, paired at most with the analysis of
h as ethnographic is misleading be-
f any ethnographic research project.
iews are elicitation methods. They
t their practices and beliefs as well
combine elicitation methods (like
tservation, which, as a method not
differences between what people
,lation is that this methodological
umption that culture is present to
t culture is something in people's
l tell the researcher, to appear later
account. Of course, persons can
a result, interviews should be part
: and other elicitation methods can
ven to ourselves. Obvious cases of
us, an insight dating back to Freud.
1honological rule like assimilation,
tes m in impossible because p is a
al n assimilates to this place of ar-
this rule in an interview, even
In the flow of everyday speech.
ied to language and the psyche. In
show how much of everyday so-
rdieu emphasized this point when
ing' a culture: 'it is the analogy
around in a foreign landscape'
part of your daily routine. If there
how many stairs are there? The
ledge via an interview, where the
of elicitation 'inevitably induced
sa result,
gly for his own use the representa-
hen they lack practical mastery of a
selves with an explicit and at least
ertoire of rules. (Bourdieu 1977: 2)
Rethinking Digital Anthropology 55
Elicitation not interwoven with participant observation can lead researchers to con-
fuse representation with reality, and thereby mistakenly equate culture with rules,
scripts or norms rather than embodied practices.
If there is one thing that ethnographers have shown over the years, it is that 'what
is essential goes without saying because it comes without saying: the tradition is si-
lent, not least about itself as a tradition' (Bourdieu 1977: 167, emphasis in original).
When ethnographers ask interview questions, they obtain representations of social
practice. Representations are certainly social facts (Rabinow 1986) and have cultural
effects. But they cannot be conflated with culture as a whole. If you ask someone
'what does friendship mean to you?' you will get a representation of what that person
takes friendship to be. That representation is socially consequential; it is embedded
in (and influences) a cultural context. However, that elicited representation is not
identical to friendship in practice.
The methodological contribution of participant observation is that it provides eth-
nographers insight into practices and meanings as they unfold. It also allows for
obtaining nonelicited data--conversations as they occur, but also activities, embodi-
ments, movements though space, and built environments. For instance in Part 2, I
observed Second Life residents teaching each other how to skate on a virtual ice rink,
in part by learning how to skate myself. Had I just walked up to an avatar and asked
out of the blue, 'how do you learn in Second Life?' I would have likely received a
formal response emphasizing things traditionally seen as learning-related; rich de-
tail about a group of avatars learning to skate would not have been in the offing.
Participant observation allows researchers to identify cultural practices and beliefs
ofwhich they were unaware during the process of research design.
Some persons terming themselves ethnographers may not wish to hear this. On
more than one occasion I have counselled scholars who claim to be 'doing ethnog-
raphy' but use interviews in isolation-in one case, because a colleague told the
scholar that participant observation would take too long. Participant observation is
never rapid: 'not unlike learning another language, such inquiry requires time and
patience. There are no shortcuts' (Rosaldo 1989: 25). You cannot become fluent in
a new language overnight, or even in a month or two. Similarly, someone claiming
to have conducted ethnographic research in a week or even a month is mischaracter-
izing his or her work unless it is part of a more long-term engagement. There is no
way the researcher could have become known to a community and participated in its
everyday practices in such a compressed time frame.
Conclusion: Time and Imagination
When I think about the exciting possibilities that inhere in rethinking digital anthro-
pology, I find my mind wandering back to an image. A webpage, to be precise, that
has haunted me for years despite its apparent triviality. I think-of all things!-about
the original McDonald's home page from 1996, from the early days of the Internet's
56 Digital Anthropology
ascendance.
16
Despite its simplicity from a contemporary perspective (basically, the
Golden Arches logo on a red background), the webpage represented the best that a
major corporation could offer in terms of web presence; it likely involved consider-
able expense to design and implement.
When I think about what this website represents, I compare it to some contempor-
ary phenomenon like Facebook or Twitter. For instance, the well-known microblog-
ging site Twitter was founded in 2006 and allows users to post text messages up to 140
characters in length. Such sites are simple; broadband Internet connections and blaz-
ing graphics cards are unnecessary for their operation. One could effectively access
Twitter with a slow dial-up connection, using a 1990s-era computer with what would
now be minuscule processing power. In fact, there is no technological reason why
Twitter could not have existed in 1996, alongside that original McDonald's home page.
Why did Twitter not exist in 1996, coming into being only ten years later? It was
not a limit of technology; it was a limit of imagination. In the early years of widespread
web connectivity, we did not yet realize the affordances of the technology in question.
Virtual worlds, online games, social networking sites and even instant messaging
and smartphones in the 201 Os are analogous to that McDonald's webpage from 1996.
Current uses of these technologies push against the horizon of the familiar, and it
could not be otherwise. Transformative potential uses of these technologies certainly
exist, but at present they are no more conceivable than the idea of a Twitter feed
would have been to a user of the McDonald's website in 1996, despite its feasibility
from a technical standpoint. It is a matter of time and imagination.
Leach concluded 'Rethinking Anthropology' by emphasizing: 'I believe that we
social anthropologists are like the mediaeval Ptolemaic astronomers; we spend our
time trying to fit the facts of the objective world into the framework of a set of con-
cepts which have been developed a priori instead of from observation' (Leach 1961:
26). Leach was frustrated that social researchers often fail to listen to the empirical
realities they ostensibly study. Despite our best intentions, we often fall back on folk
theories and preconceived notions from our own cultural backgrounds. This is par-
ticularly the case when speaking about the future. The problem with the future is that
there is no way to research it. It is the domain of the science fiction author and the
entrepreneur on the make. Social scientists study the past, and many of them, includ-
ing ethnographers, study the present; in this chapter I have worked to demonstrate
how digital anthropology might contribute to studying this emergent present. But if
we see that contribution as showing that the virtual and actual are no longer separate,
we will have substituted a mistaken teleology for empirical reality: we will remain
in a Ptolemaic frame of mind.
The virtual and the actual are not blurring, nor are they pulling apart from one
another. Such spatial metaphors of proximity and movement radically mischaracter-
ize the semiotic and material interchanges that forge both the virtual and the actual.
Digital anthropology as a framework can provide tools to avoid this conceptual cul-
de-sac-via a theoretical attention to the indexical relationships that link the online
and offline through similitude and dijferen1
ticipant observation.
Social researchers are constantly asked
'trending' to predict what will happen witl
access to a time machine and confronted 1
to predict even the rise of blogging, our 01
gating the past and present. Digital anthro
regard, but for this to happen it must stan<
is a necessity for digital anthropology-yc
weekend. But imagination is also needed.
short if it does not include imagining wha1
quences might be for social inquiry.
Notes
I thank Daniel Miller and Heather Horst f
ter and Paul Manning for his helpful coffil
1. In this chapter I treat actual, physica
onyms. It is possible to craft frame\
a flawed folk theory of language th
entails multiple corresponding entiti
2. I have briefly discussed these meani
embodiment (Boellstorff 2011: 514-
3. For reviews of the history of digital:
storff(2008: chap. 2); Boellstorff, N
Coleman (2010).
4. For example Curtis ([ 1992] 1997), I<
( 1991 ). Such uses of convergence d
of convergence culture, which refer'
5. Lehdonvirta used the unwieldy ph
and virtual environments (MMO[sJ:
6. This is true as well with regard to
derstood notion of the 'magic circle
2008: 23).
7. Of course, many religious traditio
virtual (as exemplified by the nc
However, the Christian tradition 1
Western contexts, where the Intern
205-11).
8. In their introduction to this volume
rethink basic anthropological ideas
:t contemporary perspective (basically, the
), the webpage represented the best that a
web presence; it likely involved consider-
I compare it to some contempor-
. For instance, the well-known microblog-
Uows users to post text messages up to 140
broadband Internet connections and blaz-
ir operation. One could effectively access
ng a 1990s-era computer with what would
act, there is no technological reason why
gside that original McDonald's home page.
1ing into being only ten years later? It was
1agination. In the early years of widespread
affordances of the technology in question.
;working sites and even instant messaging
.s to that McDonald's webpage from 1996.
tgainst the horizon of the familiar, and it
tential uses of these technologies certainly
tlceivable than the idea of a Twitter feed
.d's website in 1996, despite its feasibility
>ftime and imagination.
>logy' by emphasizing: 'I believe that we
val Ptolemaic astronomers; we spend our
world into the framework of a set of con-
nstead of from observation' (Leach 1961 :
rchers often fail to listen to the empirical
best intentions, we often fall back on folk
:rr own cultural backgrounds. This is par-
future. The problem with the future is that
pain of the science fiction author and the
1study the past, and many of them, includ-
'is chapter I have worked to demonstrate
I to studying this emergent present. But if
1
virtual and actual are no longer separate,
gy for empirical reality: we will remain
'ng, nor are they pulling apart from one
ty and movement radically mischaracter-
hat forge both the virtual and the actual.
ovide tools to avoid this conceptual cul-
dexical relationships that link the online
Rethinking Digital Anthropology 57
and offline through similitude and difference and by a methodological focus on par-
ticipant observation.
Social researchers are constantly asked to engage in the work of forecasting or
'trending' to predict what will happen with regard to new technologies. But lacking
access to a time machine and confronted by the failure of the most savvy futurists
to predict even the rise ofblogging, our only real explanatory power lies in investi-
gating the past and present. Digital anthropology can play an important role in this
regard, but for this to happen it must stand for more than ethnography online. Time
is a necessity for digital anthropology-you cannot do ethnographic research over a
weekend. But imagination is also needed. Rethinking digital anthropology will fall
short if it does not include imagining what, 'digital' might mean and what its conse-
quences might be for social inquiry.
Notes
I thank Daniel Miller and Heather Horst for their encouragement to write this chap-
ter and Paul Manning for his helpful comments.
1. In this chapter I treat actual, physical and offline and virtual and online as syn-
onyms. It is possible to craft frameworks in which these terms differ, but it is
a flawed folk theory of language that the mere existence of multiple lexemes
entails multiple corresponding entities in the world.
2. I have briefly discussed these meanings of the digital elsewhere with regard to
embodiment (Boellstorff 20 II: 514-15).
3. For reviews of the history of digital anthropological work, see, inter alia, Boell-
storff(2008: chap. 2); Boellstorff, Nardi, Pearce and Taylor (2012: chap. 2) and
Coleman (2010).
4. For example Curtis ([1992] 1997), Kendall (2002) and Morningstar and Farmer
(1991). Such uses of convergence diverge from Henry Jenkins's (2008) notion
of convergence culture, which references differing media.
5. Lehdonvirta used the unwieldy phrase 'massively-multiplayer online games
and virtual environments (MMO[s])'; I will simply use 'virtual worlds' here.
6. This is true as well with regard to Huizinga's much-maligned and poorly un-
derstood notion ofthe 'magic circle' (Huizinga [1938) 1950: 57; see Boellstorff
2008: 23).
7. Of course, many religious traditions have influenced understandings of the
virtual (as exemplified by the notion of avatars, drawn from Hinduism).
However, the Christian tradition has dominated, given its hegemony in the
Western contexts, where the Internet revolution began. See Boellstorff (2008:
205-11).
8. In their introduction to this volume, Miller and Horst also speak of the need to
rethink basic anthropological ideas in light of the impact of the digital.
58 Digital Anthropology
9. Even the varied post-Saussurean approaches to language provide for the consti-
tutive role of gaps (and movement across those gaps). This includes notions of
iteration which 'contains in itselfthe discrepancy of a difference that constitutes
it as iteration' (Derrida 1988: 53, emphasis in original).
10. These debates, and my engagement with them, preceded and took place separ-
ately from debates over dubbing versus subbing that appear in some contempo-
rary debates over Internet-mediated fan production.
11. The ethnographic contexts of Indonesia and Second Life are of course very
different; the common need to challenge teleological narratives says as much
about scholarly assumptions as the contexts themselves.
12. For a detailed theoretical and methodological discussion of this research, see
Boellstorff (2008) and Boellstorff et al. (2012).
13. In English and many other languages (for example Indonesian), speakers use
lexical items like sir or madam to optionally index intimacy. For a discussion of
social indexicality and social deixis more generally, see Manning (2001).
14. What was likely the first contemporary virtual world originated in two hands
pointing at each other while superimposed on a computer screen (Krueger
1983; see Boellstorff2008: 42-7).
15. Phrases such as 'digital archaeology' usually connote a historical approach
rather than a true engagement with archaeological approaches and paradigms
(for one notable exception, see Jones 1997).
16. You can see this webpage at http:l/web.archive.org/web/19961221230104/
http:/www.mcdonalds.com/.
References
Bedos-Rezak, B. 2000. Medieval Identity: A Sign and a Concept. American His-
torical Review 1 05(5). http://www.historycooperative.org/journals/ahr/l 05.5/
ah001489.html.
Boellstorff, T. 2005. The Gay Archipelago: Sexuality and Nation in Indonesia. Prince-
ton, NJ: Princeton University Press.
Boellstorff, T. 2008. Coming of Age in Second Life: An Anthropologist Explores the
Virtually Human. Princeton, NJ: Princeton University Press.
Boellstorff, T. 2011. Placing the Virtual Body: Avatar, Chora, Cypherg. In A Com-
panion to the Anthropology of the Body and Embodiment, ed. F. E. Mascia-Lees,
504-20. New York: Wiley-Blackwell.
Boellstorff, T., B. A. Nardi, C. Pearce and T. L. Taylor. 2012. Ethnography and Vir-
tual Worlds: A Handbook of Method. Princeton, NJ: Princeton University Press.
Bourdieu, P. 1977. Outline of a Theory of Practice. Cambridge: Cambridge Univer-
sity Press.
Coleman, E. G. 2010. Ethnographic Ap
Anthropology 39: 487-505.
Curtis, P. [1992] 1997. Mudding: Soc
ties. In Culture of the Internet, ed.
Erlbaum.
Derrida, J. 1988. Signature Event Conte
western University Press.
Duranti, A. 1997. Linguistic Anthropolo,
Geertz, C. 1983. 'From the Native's Poi
Understanding. In Local Knowledg
ogy, 55-72. New York: Basic Book
Gershon, I. 2010a. Breaking Up Is Har
gies. Journal of Linguistic Anthropt
Gershon, I. 201 Ob. The Breakup 2. 0:
Cornell University Press.
Hanks, W. F. 1992. The Indexical GfOI
text: Language as an Interactive P
43-76. Cambridge: Cambridge Uni
Huizinga, J. [1938] 1950. Homo
Boston: Beacon Press.
Inoue, M. 2003. Speech without a Spe:
Translation. Language and Commu
Jenkins, H. 2008. Convergence Cultu.
York: New York University Press.
Jones, Q. 1997. Virtual-Communities,
A Theoretical Outline. Journal ofO
onlinelibrary.wiley.com/doi/1 0.111
April 12, 2012).
Keane, W. 2003. Semiotics and the S
and Communication 23(3/4): 409-:
Kendall, L. 2002. Hanging Out in the
Online. Berkeley: University of Ca
Krueger, M. W. 1983. Artificial
Leach, E. R. 1961. Rethinking Anth!
London: Robert Cunningham and
Lehdonvirta, V. 2010. Virtual World:
Approach in MMO Studies.
10(1): 1-16.
Levinson, S C. 1983. Pragmatics. Cat
Malinowski, B. 1922. Argonauts ofth
Manning, H. P. 2001. On Social Deixi
hes to language provide for the consti-
; those gaps). This includes notions of
of a difference that constitutes
is in original).
them, preceded and took place separ-
llbbing that appear in some contempo-
troduction.
and Second Life are of course very
: teleological narratives says as much
K.ts themselves.
>gical discussion of this research, see
2012).
Jr example Indonesian), speakers use
.lly index intimacy. For a discussion of
:generally, see Manning (2001).
virtual world originated in two hands
>sed on a computer screen (Krueger
sually connote a historical approach
aeological approaches and paradigms
'7).
eb.archive.org/web/19961221230 104/
1 Sign and a Concept. American His-
05.5/
'uality and Nation in Indonesia. Prince-
!
f Life: An Anthropologist Explores the
'university Press.
: Avatar, Chora, Cypherg. In A Com-
Embodiment, ed. F. E. Mascia-Lees,
. Taylor. 2012. Ethnography and Vir-
ton, NJ: Princeton University Press.
tice. Cambridge: Cambridge Univer-
Rethinking Digital Anthropology 59
Coleman, E. G. 2010. Ethnographic Approaches to Digital Media. Annual Review of
Anthropology 39: 487-505.
Curtis, P. [1992] 1997. Mudding: Social Phenomena in Text-Based Virtual Reali-
ties. In Culture ofthe Internet, ed. S. Kiesler, 121--42. Mahwah, NJ: Lawrence
Erlbaum.
Derrida, J. 1988. Signature Event Context. In Limited Inc., 1-24. Evanston, IL: North-
western University Press.
Duranti, A. 1997. Linguistic Anthropology. Cambridge: Cambridge University Press.
Geertz, C. 1983. 'From the Native's PointofView': On the Nature of Anthropological
Understanding. In Local Knowledge: Further Essays in Interpretive Anthropol-
ogy, 55-72. New York: Basic Books.
Gershon, I. 201 Oa. Breaking Up Is Hard To Do: Media Switching and Media Ideolo-
gies. Journal of Linguistic Anthropology 20(2): 389--405 .
Gershon, I. 2010b. The Breakup 2.0: Disconnecting over New Media. Ithaca, NY:
Cornell University Press.
Hanks, W. F. 1992. The Indexical Ground of Deictic Reference. In Rethinking Con-
text: Language as an Interactive Phenomenon, ed. C. Goodwin and A. Duranti,
43-76. Cambridge: Cambridge University Press.
Huizinga, J. [1938] 1950. Homo Ludens: A Study of the Play-Element in Culture.
Boston: Beacon Press.
Inoue, M. 2003. Speech without a Speaking Body: 'Japanese Women's Language' in
Translation. Language and Communication 23(3/4): 315-30.
Jenkins, H. 2008. Convergence Culture: Where Old and New Media Collide. New
York: New York University Press.
Jones, Q. 1997. Virtual-Communities, Virtual Settlements, and Cyber-Archae-ology:
A Theoretical Outline. Journal of Computer-Mediated Communication 3(3). http:/ I
onlinelibrary.wiley.corn/doi/1 0.1111/j.l 083-61 01.1997.tb00075.x/full (accessed
April 12, 2012).
Keane, W. 2003. Semiotics and the Social Analysis of Material Things. Language
and Communication 23(3/4): 409-25.
Kendall, L. 2002. Hanging Out in the Virtual Pub: Masculinities and Relationships
Online. Berkeley: University of California Press.
Krueger, M. W. 1983. Artificial Reality. Reading, MA: Addison-Wesley.
Leach, E. R. 1961. Rethinking Anthropology. In Rethinking Anthropology, 1-27.
London: Robert Cunningham and Sons.
Lehdonvirta, V. 2010. Virtual Worlds Don't Exist: Questioning the Dichotomous
Approach in MMO Studies. International Journal of Computer Game Research
10(1): 1-16.
Levinson, S C. 1983. Pragmatics. Cambridge: Cambridge University Press.
Malinowski, B. 1922. Argonauts of the Western Pacific. New York: E. P. Dutton.
Manning, H. P. 2001. On Social Deixis. Anthropological Linguistics 43: 54-100.
60 Digital Anthropology
Marcus, G. 1995. Ethnography in/of the World System: The Emergence of Multi-
Sited Ethnography. Annual Review of Anthropology 24: 95-117.
Miller, D. 2011. Tales from Face book. Cambridge: Polity Press.
Morningstar, C., and F. R. Farmer. 1991. The Lessons of Lucasfilm's Habitat. In
Cyberspace: First Steps, ed. M. Benedikt, 273-301. Cambridge, MA: MIT Press.
Oetomo, D. 1997. Ketika Sharon Stone Berbahasa Indonesia [When Sharon Stone
speaks Indonesian]. In Bercinta Dengan Televisi: llusi, lmpresi, dan Jmaji Sebuah
Kotak Ajaib [In love with television: Illusions, impressions, and images from a
magical box], ed. D. Mulyana and I. Subandy Ibrahim, 333-7. Bandung, Indone-
sia: PT Remaja Rosdakarya.
Pearce, C., and Artemesia. 2009. Communities of Play: Emergent Cultures in Multi-
player Games and Virtual Worlds. Cambridge, MA: MIT Press.
Rabinow, P. 1986. Representations Are Social Facts: Modernity and Post-Modernity
in Anthropology. In Writing Culture: the Poetics and Politics of Ethnography, ed.
J. Clifford and G. E. Marcus, 234-61. Berkeley: University of California Press.
Rogers, R. 2009. The End of the Virtual: Digital Methods. Amsterdam: Vossiuspers
UvA.
Rosaldo, R. 1989. Culture and Truth: The Remaking of Social Analysis. Boston:
Beacon Press.
Sahlins, M. 1996. The Sadness of Sweetness: The Native Anthropology of Western
Cosmology. Current Anthropology 37(3): 395--428.
Silverstein, M. 1976. Shifters, Linguistic Categories, and Cultural Description. In
Meaning in Anthropology, ed. K. H. Basso and H. A. Selby, 11-55. Albuquerque:
University ofNew Mexico Press.
Strathern, M. 2004. Commons and Borderlands: Working Papers on lnterdisciplin-
arity, Accountability, and the Flow of Knowledge. Wantage: Sean Kingston.
Domestic
ogy. Begit
of the Am
house soci
ism of the
anthropol<
tion andn
has broad1
or non-W
t t e n t i o n ~
cesses un1
and social
1984) anc
restructur1
which ge1
domestic
Patterr
balization
domestic
altemativ1
and dwell
and neoli
1
notions o
devices, 1
1999; Ha
sumption
(Cieraad
plays in 1
Garvey 2
This chat
the home